Validity and accuracy of regional bioelectrical impedance devices to determine whole-body fatness.

OBJECTIVE Growing emphasis on obesity as a risk factor for chronic diseases and commercial availability of impedance devices for the at-home assessment of body fatness have stimulated the need for a critical evaluation of the validity of these instruments. This study determined the reproducibility and accuracy of two commercial impedance devices that use upper (hand-to-hand) or lower (foot-to-foot) body contact electrode placements in adults with a wide range of body fatness. METHODS Body composition was assessed with dual x-ray absorptiometry in apparently healthy adults (62 women and 48 men) ages 21 to 60 y, with a range in body mass index of 18.6 to 40.5 kg/m2. Variability in body fatness predicted with the regional body impedance devices was determined in 10 adults on 5 consecutive d. A 50-kHz, tetrapolar bioelectrical impedance plethysmograph with surface electrode placements on the upper and lower limbs was used to determine reference regional and whole-body impedance values. RESULTS Variability in body mass (1%) over 5 d was less than body fatness predicted with the upper (2-10%) and lower (3-5%) body devices. Regional and whole-body impedance values were different (P < 0.05) in the women, whereas upper and lower body values were lower (P < 0.05) than whole-body impedance in the men. Dual x-ray absorptiometric determinations of body fatness were similar to predictions based on models derived from physical characteristics (age, stature, body mass, and sex) but significantly different (P < 0.05) from estimates from the impedance devices, which underestimated body fatness. Bias in predictions of body fatness with the regional devices was systematically (P < 0.0001) related to body fatness. CONCLUSION Use of regional impedance devices to assess body fatness is limited by a lack of precision and accuracy.